[SERVER-26748] Don't call for priority takeover when repl lagged Created: 24/Oct/16 Updated: 25/Jan/17 Resolved: 01/Nov/16 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Replication |
| Affects Version/s: | None |
| Fix Version/s: | 3.2.11, 3.4.0-rc3 |
| Type: | Improvement | Priority: | Major - P3 |
| Reporter: | Spencer Brody (Inactive) | Assignee: | Spencer Brody (Inactive) |
| Resolution: | Done | Votes: | 0 |
| Labels: | code-and-test | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||||||||||||||
| Backwards Compatibility: | Minor Change | ||||||||||||||||||||||||
| Backport Completed: | |||||||||||||||||||||||||
| Sprint: | Repl 2016-11-21 | ||||||||||||||||||||||||
| Participants: | |||||||||||||||||||||||||
| Description |
|
Currently in pv1 we call for priority takeover without concern for whether the higher priority node is caught up in replication, which can lead to unnecessary rollbacks of w:1 writes. If we avoid calling for an election due to priority takeover if we are more than X seconds repl lagged, then that in combination with the catchup period should prevent any writes from being lost during priority takeover. |
| Comments |
| Comment by Githook User [ 07/Nov/16 ] |
|
Author: {u'username': u'stbrody', u'name': u'Spencer T Brody', u'email': u'spencer@mongodb.com'}Message: Backport of 286a682a5ecff2e1646cbe68315529f7b3f6bd7c |
| Comment by Spencer Brody (Inactive) [ 01/Nov/16 ] |
|
This work introduced a new startup-only setParameter, priorityTakeoverFreshnessWindowSeconds, which controls how up-to-date a secondary must be to call for priority takeover. |
| Comment by Githook User [ 01/Nov/16 ] |
|
Author: {u'username': u'stbrody', u'name': u'Spencer T Brody', u'email': u'spencer@mongodb.com'}Message: |
| Comment by Githook User [ 01/Nov/16 ] |
|
Author: {u'username': u'stbrody', u'name': u'Spencer T Brody', u'email': u'spencer@mongodb.com'}Message: |